CN116773894B - Collector power failure detection system and method thereof - Google Patents
Collector power failure detection system and method thereof Download PDFInfo
- Publication number
- CN116773894B CN116773894B CN202310710148.3A CN202310710148A CN116773894B CN 116773894 B CN116773894 B CN 116773894B CN 202310710148 A CN202310710148 A CN 202310710148A CN 116773894 B CN116773894 B CN 116773894B
- Authority
- CN
- China
- Prior art keywords
- waveform
- sampling window
- voltage
- voltage sampling
- context
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000001514 detection method Methods 0.000 title claims abstract description 46
- 238000000034 method Methods 0.000 title description 29
- 238000005070 sampling Methods 0.000 claims description 175
- 239000013598 vector Substances 0.000 claims description 139
- 238000012545 processing Methods 0.000 claims description 37
- 230000009467 reduction Effects 0.000 claims description 24
- 238000009826 distribution Methods 0.000 claims description 21
- 238000005457 optimization Methods 0.000 claims description 21
- 238000000605 extraction Methods 0.000 claims description 20
- 238000013527 convolutional neural network Methods 0.000 claims description 15
- 230000006870 function Effects 0.000 claims description 15
- 230000004927 fusion Effects 0.000 claims description 14
- 239000011159 matrix material Substances 0.000 claims description 11
- 238000011156 evaluation Methods 0.000 claims description 10
- 238000012795 verification Methods 0.000 claims description 10
- 230000004913 activation Effects 0.000 claims description 2
- 238000011176 pooling Methods 0.000 claims description 2
- 238000012546 transfer Methods 0.000 claims description 2
- 238000012360 testing method Methods 0.000 claims 1
- 230000002159 abnormal effect Effects 0.000 description 12
- 238000004422 calculation algorithm Methods 0.000 description 9
- 238000010586 diagram Methods 0.000 description 9
- 230000008569 process Effects 0.000 description 9
- 238000004458 analytical method Methods 0.000 description 6
- 238000004590 computer program Methods 0.000 description 6
- 230000000694 effects Effects 0.000 description 6
- 238000007781 pre-processing Methods 0.000 description 6
- 238000005314 correlation function Methods 0.000 description 4
- 230000009466 transformation Effects 0.000 description 4
- 238000004891 communication Methods 0.000 description 3
- 230000008878 coupling Effects 0.000 description 3
- 238000010168 coupling process Methods 0.000 description 3
- 238000005859 coupling reaction Methods 0.000 description 3
- 238000003745 diagnosis Methods 0.000 description 3
- 230000002411 adverse Effects 0.000 description 2
- 230000032683 aging Effects 0.000 description 2
- 238000013528 artificial neural network Methods 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 238000000354 decomposition reaction Methods 0.000 description 2
- 238000013135 deep learning Methods 0.000 description 2
- 238000001914 filtration Methods 0.000 description 2
- 238000007726 management method Methods 0.000 description 2
- 238000005065 mining Methods 0.000 description 2
- 230000008054 signal transmission Effects 0.000 description 2
- 238000007476 Maximum Likelihood Methods 0.000 description 1
- 230000005856 abnormality Effects 0.000 description 1
- 230000003321 amplification Effects 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000012512 characterization method Methods 0.000 description 1
- 230000000295 complement effect Effects 0.000 description 1
- 238000005094 computer simulation Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000013480 data collection Methods 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 230000003111 delayed effect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 239000000835 fiber Substances 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 238000012067 mathematical method Methods 0.000 description 1
- 238000003058 natural language processing Methods 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 238000003199 nucleic acid amplification method Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000008439 repair process Effects 0.000 description 1
- 230000003252 repetitive effect Effects 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 230000000630 rising effect Effects 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 230000035945 sensitivity Effects 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 230000003595 spectral effect Effects 0.000 description 1
- 238000001228 spectrum Methods 0.000 description 1
- 238000003786 synthesis reaction Methods 0.000 description 1
- 238000012549 training Methods 0.000 description 1
- 230000001052 transient effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R19/00—Arrangements for measuring currents or voltages or for indicating presence or sign thereof
- G01R19/165—Indicating that current or voltage is either above or below a predetermined value or within or outside a predetermined range of values
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
- G01R31/50—Testing of electric apparatus, lines, cables or components for short-circuits, continuity, leakage current or incorrect line connections
- G01R31/54—Testing for continuity
Abstract
The application relates to the technical field of power failure detection, and particularly discloses a power failure detection system and a power failure detection method of a collector.
Description
Technical Field
The application relates to the technical field of power failure detection, in particular to a power failure detection system and a power failure detection method of a collector.
Background
The collector is an automatic device with the functions of on-site real-time data collection and processing, has the functions of real-time collection, automatic storage, instant display, instant feedback, automatic processing, automatic transmission and the like, and provides a guarantee for the authenticity, effectiveness, real-time performance and usability of on-site data.
In the working process of the collector, the low-voltage side power supply supplies power to the battery system, and if the low-voltage side power supply is accidentally powered down, the battery system can be powered down accidentally. At present, in the actual operation process of the collector, vibration of equipment or aging of a wire harness connector in the equipment may cause accidental disconnection of a low-voltage side power supply in the collector, so that a battery system is powered down accidentally. Unexpected power down of the battery system has an adverse effect on the collector, for example, unexpected conduction or unexpected disconnection of the contactor may be caused, so that the contactor is damaged, and the operation of the collector is abnormal.
Accordingly, an optimized collector power down detection system is desired.
Disclosure of Invention
The application provides a power failure detection system and a power failure detection method for a collector, which can accurately detect whether a power failure event occurs to a low-voltage side power supply, thereby ensuring the normal operation of the collector and improving the safety and stability of the operation of the collector.
In a first aspect, a collector power down detection system is provided, the system comprising: the signal acquisition module is used for acquiring an output voltage signal output by the low-voltage side power supply; the signal noise reduction module is used for carrying out noise reduction processing on the output voltage signal to obtain a noise-reduced output voltage signal; the sampling module is used for carrying out sliding window type sampling based on the sampling window on the noise-reduced output voltage signal so as to obtain a plurality of output voltage local sampling window signals; the voltage waveform characteristic extraction module is used for respectively passing the plurality of output voltage local sampling window signals through a voltage waveform characteristic extractor based on a convolutional neural network model to obtain a plurality of voltage sampling window waveform characteristic vectors; the waveform characteristic global correlation module is used for enabling the waveform characteristic vectors of the voltage sampling windows to pass through a waveform inter-correlation characteristic extractor based on a converter model to obtain classification characteristic vectors; and the power-down detection module is used for enabling the classification feature vector to pass through a classifier to obtain a classification result, wherein the classification result is used for indicating whether a power-down event occurs or not.
In a second aspect, a method for detecting power failure of a collector is provided, where the method includes: obtaining an output voltage signal output by a low-voltage side power supply; carrying out noise reduction treatment on the output voltage signal to obtain a noise-reduced output voltage signal; sliding window type sampling based on sampling windows is carried out on the noise-reduced output voltage signals so as to obtain a plurality of output voltage local sampling window signals; the output voltage local sampling window signals are respectively passed through a voltage waveform characteristic extractor based on a convolutional neural network model to obtain a plurality of voltage sampling window waveform characteristic vectors; passing the plurality of voltage sampling window waveform feature vectors through a converter model-based inter-waveform correlation feature extractor to obtain classification feature vectors; and passing the classification feature vector through a classifier to obtain a classification result, wherein the classification result is used for indicating whether a power failure event occurs.
In a third aspect, there is provided a chip comprising an input-output interface, at least one processor, at least one memory and a bus, the at least one memory to store instructions, the at least one processor to invoke the instructions in the at least one memory to perform the method in the second aspect.
In a fourth aspect, a computer readable medium is provided for storing a computer program comprising instructions for performing the method of the second aspect described above.
In a fifth aspect, there is provided a computer program product comprising instructions which, when executed by a computer, perform the method of the second aspect described above.
The collector power failure detection system and the collector power failure detection method can accurately detect whether a power failure event occurs to a low-voltage side power supply, so that normal operation of the collector is ensured, and safety and stability of operation of the collector are improved.
Drawings
Fig. 1 is a schematic block diagram of a collector power down detection system according to an embodiment of the application.
Fig. 2 is a schematic structural diagram of a global association module of waveform characteristics in a power-down detection system of a collector according to an embodiment of the present application.
Fig. 3 is a schematic block diagram of a waveform correlation feature extraction unit in the collector power failure detection system according to the embodiment of the present application.
Fig. 4 is a schematic flow chart of a collector power failure detection method according to an embodiment of the present application.
Fig. 5 is a schematic diagram of a model architecture of a collector power failure detection method according to an embodiment of the present application.
Detailed Description
The following description of the embodiments of the present application will be made clearly and fully with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some, but not all embodiments of the application. All other embodiments, which can be made by those skilled in the art based on the embodiments of the application without making any inventive effort, are also within the scope of the application.
As described above, at present, during the actual operation of the harvester, vibration of the device or aging of the harness connector in the device may cause the low-voltage side power supply in the harvester to be disconnected accidentally, thereby causing the battery system to be powered down accidentally. Unexpected power down of the battery system has an adverse effect on the collector, for example, unexpected conduction or unexpected disconnection of the contactor may be caused, so that the contactor is damaged, and the operation of the collector is abnormal. Accordingly, an optimized collector power down detection system is desired.
Accordingly, in order to enable power failure detection of the collector in the working process of the collector, whether a power failure event occurs to the low-voltage side power supply or not is expected to be determined based on analysis of an output voltage signal of the low-voltage side power supply, so that normal working of the collector is guaranteed, and safety and stability of working of the collector are improved. However, a great amount of noise interference exists in the process of collecting the output voltage signal output by the low-voltage side power supply, so that the characteristics of the output voltage signal become fuzzy, and the detection and judgment of the power failure event are affected. And, also consider because the implicit characteristic of the information about voltage stability and abnormal characteristic in the said output voltage signal is small-scale, it is difficult to carry on the effective capture and characterization through the traditional characteristic extraction mode. Therefore, in this process, the difficulty lies in how to fully express the stability characteristics and abnormal characteristic information about the output voltage in the output voltage signal, so as to accurately detect whether the low-voltage side power supply has a power failure event, thereby ensuring the normal operation of the collector and improving the safety and stability of the operation of the collector.
In recent years, deep learning and neural networks have been widely used in the fields of computer vision, natural language processing, text signal processing, and the like. The development of deep learning and neural networks provides new solutions and solutions for mining stability characteristics and abnormal characteristic information about output voltages in the output voltage signals.
Fig. 1 is a schematic block diagram of a collector power down detection system according to an embodiment of the application. As shown in fig. 1, the collector power down detection system 100 includes:
the signal acquisition module 110 is configured to acquire an output voltage signal output by the low-voltage side power supply. It should be understood that the low side power output refers to the power output by the power system on the low side. In an electric power system, a voltage is generally divided into a high-voltage side and a low-voltage side, and a low-voltage side power supply output refers to electric energy output on the low-voltage side. The low-side power supply is typically referred to as a dc power supply, whose output voltage is typically between a few volts and a few tens of volts, for powering the battery management system. Specifically, in one embodiment of the present application, the output voltage signal of the low side power supply output is collected by a multimeter or digital voltmeter. Specifically, first, the low-voltage side power supply output terminal of the collector is found. Then, the positive electrode of the voltmeter is connected to the output terminal, and the negative electrode of the voltmeter is connected to the ground terminal of the collector.
The signal noise reduction module 120 is configured to perform noise reduction processing on the output voltage signal to obtain a noise-reduced output voltage signal. It should be appreciated that during the actual acquisition of the output voltage signal, the output voltage signal may be subject to noise, clutter, etc., which may affect the accuracy and precision of the output voltage signal. In addition, during signal transmission, the noise interference may gradually accumulate, resulting in a lower signal-to-noise ratio, and eventually may make it difficult to accurately decode the signal. Therefore, when the output voltage signal is acquired, noise reduction processing is required for the signal. Based on the above, in the technical scheme of the application, the noise reduction processing is further performed on the output voltage signal to obtain a noise-reduced output voltage signal. By performing noise reduction processing on the output voltage signal, an interference signal irrelevant to an original signal can be filtered, so that a cleaner and purer output voltage signal is obtained. Therefore, the subsequent signal processing and analysis work can be more effective and accurate, and the data quality and the reliability are improved.
Optionally, in an embodiment of the present application, the signal noise reduction module 120 performs noise reduction processing on the output voltage signal using a dual-tree complex wavelet transform to obtain a noise-reduced output voltage signal. It should be understood that discrete wavelet transformation is to divide an input signal into a high frequency component and a low frequency component by a high pass filter and a low pass filter, and then obtain wavelet coefficients by downsampling. However, wavelet decomposition has problems of poor translational sensitivity, lack of directional selectivity, and the like, and seriously affects the ability of the wavelet to represent the original signal. To overcome the drawbacks of wavelet decomposition, a dual-tree complex wavelet transform is proposed. The dual-tree complex wavelet transformation has the advantages of translational invariance, direction selectivity, high-efficiency calculation efficiency, good reconfigurability and the like, and has been widely applied to noise reduction processing.
Optionally, in another embodiment of the present application, the signal noise reduction module 120 includes: and the signal preprocessing unit is used for preprocessing the acquired output voltage signal to obtain a preprocessed output voltage signal. It should be appreciated that the influence of noise can be reduced by the preprocessing method, facilitating subsequent processing. Specifically, in a sub-example of this embodiment, the method of preprocessing includes: the signal is amplified by an amplifier to increase the amplitude of the signal, or filtered by a filter to remove high frequency noise. Here, it should be understood by those skilled in the art that an amplifier is an electronic device or circuit that can amplify the amplitude of an input signal to a desired level. In the preprocessing of the voltage signal output by the low-side power supply of the collector, an amplifier can be used for amplifying the amplitude of the signal so as to better perform subsequent processing. The amplifier is usually composed of one or more transistors, operational amplifiers or operational amplifiers, etc., and can amplify the voltage, current or power of the signal. The gain of the amplifier may be achieved by adjusting circuit parameters or selecting different amplifier types. The amplifier can be used in the fields of signal processing, audio amplification, radio frequency communication, power management and the like.
Specifically, in a sub-example of this embodiment, the signal is amplified using an amplifier in the following formula; wherein the formula is
V out ＝G(V in -V offset )
Wherein V is in For obtaining the output voltage signal V offset G is the gain of the amplifier, V out To output a voltage signal after preprocessing. In the above formula, the original signal is first subjected to offset processing, and the zero point of the signal is shifted to the operating range of the amplifier. The signal is then amplified using an amplifier to increase the amplitude of the signal. Finally, the preprocessed voltage signal is obtained.
And the signal noise reduction unit is used for carrying out noise reduction processing on the signal by using a noise reduction algorithm on the preprocessed signal. It should be appreciated that during signal acquisition and processing, various noise often exists in the signal due to environmental noise, electromagnetic interference, signal transmission, and the like. These noise can affect the quality and accuracy of the signal, reducing the reliability and usability of the signal. Therefore, noise reduction processing is required for the pre-processed signal to improve the quality and accuracy of the signal. It will be appreciated by those skilled in the art that noise reduction algorithms are mathematical methods by which noise components of a signal can be removed by mathematical analysis and processing of the signal. Common noise reduction algorithms include wavelet transforms, wavelet packet transforms, time domain filtering, frequency domain filtering, and the like. The algorithms can select proper noise reduction methods according to the characteristics of signals and the types of noise, remove noise components in the signals and improve the signal-to-noise ratio and the reliability of the signals.
And the signal reconstruction unit is used for reconstructing the noise-reduced signal to obtain a noise-reduced output voltage signal. It should be understood that the noise-reduced signal is a digital signal that has been processed by a sampling and noise reduction algorithm, unlike the original voltage signal. Therefore, the denoised digital signal needs to be inverse transformed and restored to the original voltage signal for subsequent analysis and processing. The inverse transform algorithm is a process of converting a digital signal back to an analog signal, and commonly used inverse transform algorithms include inverse wavelet transform, inverse wavelet packet transform, inverse fourier transform, and the like. The digital signal can be converted into the same analog signal as the original signal by the inverse transformation algorithm, so that accurate restoration and analysis processing of the signal are realized.
And the sampling module 130 is configured to perform sliding window type sampling based on a sampling window on the noise-reduced output voltage signal to obtain a plurality of output voltage local sampling window signals. It should be understood that, considering that the collector has a variation rule of stability when working normally, when the low-voltage side power supply is accidentally powered down, the voltage signal is abnormal or has poor stability, and the abnormal condition may be small-scale signal characteristic information in the actual detection process. Therefore, in the technical scheme of the application, sliding window type sampling based on the sampling window is needed to be carried out on the noise-reduced output voltage signal so as to obtain a plurality of output voltage local sampling window signals. It should be appreciated that sliding window sampling is a signal sampling method that divides a continuous signal into a number of fixed length windows, with the data within each window being used as a sampling point, thereby achieving discretized sampling of the signal. Specifically, sliding window sampling continues by sliding a window of fixed length over the signal, each time taking the data within the window as a sampling point, then sliding the window forward a fixed length, continuing to sample the next point, and so on until the complete signal is sampled. The sampling method can effectively avoid the problem of aliasing in the signal sampling process and can improve the sampling efficiency to a certain extent. According to the technical scheme, the sliding window type sampling mode based on the sampling window is adopted, so that the partial sampling window signals of the output voltages can be obtained, the partial signals can reflect the characteristics and the overall trend of the original signals more carefully and comprehensively, and the change rule and the abnormal condition of the signals can be analyzed conveniently.
Optionally, in another embodiment of the present application, the noise-reduced output voltage signal is sampled by using a multi-channel sampler to obtain a plurality of output voltage local sampling window signals. It will be appreciated that the output voltage signal is sampled using an ADC (analog to digital converter) of a plurality of channels, each channel sampling the signal over a period of time and storing the sampling results in a buffer. The sampled data in the buffer may then be processed, e.g., averaged, filtered, denoised, etc., to obtain a plurality of output voltage local sampling window signals. In processing the sampled data, a Digital Signal Processor (DSP) or other similar processor may be used to achieve efficient algorithmic processing. The method can effectively improve the sampling precision and the sampling speed, and can adapt to various sampling requirements.
The voltage waveform feature extraction module 140 is configured to pass the plurality of output voltage local sampling window signals through a voltage waveform feature extractor based on a convolutional neural network model to obtain a plurality of voltage sampling window waveform feature vectors. It should be understood that, because the expression forms of the output voltage local sampling window signals in the time domain are waveform diagrams, in the technical scheme of the application, a voltage waveform feature extractor based on a convolutional neural network model with excellent expression in terms of local implicit correlation feature extraction of images is used for carrying out feature mining on the output voltage local sampling window signals, so that local implicit correlation feature information about the voltage signals in the output voltage local sampling window signals is extracted respectively, and a plurality of voltage sampling window waveform feature vectors are obtained.
Optionally, in an embodiment of the present application, the voltage waveform feature extraction module 140 is configured to: each layer of the voltage waveform characteristic extractor based on the convolutional neural network model is used for respectively carrying out input data in forward transfer of the layer: performing convolution processing on the input data based on convolution check to generate a convolution feature map; performing global average pooling processing based on a feature matrix on the convolution feature map to generate a pooled feature map; performing nonlinear activation on the feature values of all the positions in the pooled feature map to generate an activated feature map; the output of the last layer of the voltage waveform characteristic extractor based on the convolutional neural network model is the waveform characteristic vector of each voltage sampling window in the waveform characteristic vectors of the voltage sampling windows, the input of the second layer to the last layer of the voltage waveform characteristic extractor based on the convolutional neural network model is the output of the last layer, and the input of the voltage waveform characteristic extractor based on the convolutional neural network model is the output of each output voltage local sampling window in the partial sampling window signals of the output voltage.
Optionally, in another embodiment of the present application, waveform feature extraction is performed on the plurality of output voltage local sampling window signals to obtain a plurality of voltage sampling window waveform feature vectors, including: each of the plurality of output voltage local sampling window signals is feature extracted using a short-time fourier transform to obtain a feature vector. Common features include time domain features such as mean, variance, peak, etc.; frequency domain features such as frequency, power spectrum, etc.; time-frequency domain characteristics such as instantaneous frequency, instantaneous amplitude, etc. It should be appreciated that Short-time fourier transform (STFT) is a time-frequency analysis method that divides a signal into small time periods, and fourier transforms each time period to obtain frequency domain characteristics of the signal during that time period. The short-time fourier transform is used to analyze short-time frequency variations of the signal, which may provide more time-domain information than conventional fourier transforms. The main idea of short-time fourier transformation is to divide the signal into several windows, and fourier transform is performed in each window, so that a frequency domain representation of the signal in that window is obtained.
The waveform feature global correlation module 150 is configured to pass the waveform feature vectors of the plurality of voltage sampling windows through a waveform inter-correlation feature extractor based on a converter model to obtain a classification feature vector. It should be appreciated that after obtaining the local implicit characteristic about the voltage signal in the local sampling window signal of each output voltage, it is also considered that the local implicit characteristic distribution information of each output voltage has a correlation relationship about the whole output voltage signal. Therefore, in order to improve the stability of the output voltage signal output by the low-voltage side power supply and the accuracy of abnormality detection, in the technical scheme of the application, the waveform feature vectors of the voltage sampling windows are further encoded in a waveform correlation feature extractor based on a converter model so as to extract local implicit features about the voltage signal in each output voltage local sampling window signal based on the context correlation feature information of the output voltage signal overall, thereby obtaining classification feature vectors.
It should be understood that fig. 2 is a schematic structural diagram of a global association module of waveform characteristics in the power-down detection system of the collector according to an embodiment of the present application. As shown in fig. 2, the global association module 150 for waveform characteristics includes: a waveform correlation feature extraction unit 151, configured to pass the plurality of voltage sampling window waveform feature vectors through the inter-waveform correlation feature extractor based on the converter model to obtain a plurality of context voltage sampling window waveform feature vectors; the feature optimization unit 152 is configured to perform multisource information fusion pre-verification distribution evaluation optimization on the waveform feature vectors of the context voltage sampling windows to obtain waveform feature vectors of the optimized context voltage sampling windows; and the cascade fusion unit 153 is configured to cascade the optimized context voltage sampling window waveform feature vectors to obtain the classification feature vectors.
Fig. 3 is a schematic block diagram of a waveform correlation feature extraction unit in the collector power failure detection system according to the embodiment of the present application. As shown in fig. 3, the waveform-related-feature extraction unit 151 includes: a feature vector arrangement subunit 1511, configured to perform one-dimensional arrangement on the waveform feature vectors of the plurality of voltage sampling windows to obtain a waveform global feature vector; a vector-phase multiplier subunit 1512, configured to calculate a product between the global waveform feature vector and a transpose vector of each of the plurality of voltage sampling window waveform feature vectors to obtain a plurality of self-attention correlation matrices; an incidence matrix processing subunit 1513, configured to perform normalization processing on each of the plurality of self-attention incidence matrices to obtain a plurality of normalized self-attention incidence matrices; a probability unit 1514, configured to obtain a plurality of probability values by using a classification function for each normalized self-attention correlation matrix in the plurality of normalized self-attention correlation matrices; and a probability applying unit 1515, configured to weight each of the plurality of voltage sampling window waveform feature vectors with each of the plurality of probability values as a weight to obtain a plurality of context voltage sampling window waveform feature vectors.
In particular, in the technical solution of the present application, when the classification feature vector is obtained by passing the plurality of voltage sampling window waveform feature vectors through the inter-waveform correlation feature extractor based on the converter model, the classification feature vector is obtained by concatenating the plurality of context voltage sampling window waveform feature vectors obtained by passing the plurality of voltage sampling window waveform feature vectors through the inter-waveform correlation feature extractor based on the converter model, and therefore, the plurality of context voltage sampling window waveform feature vectors may be regarded as a local feature set in the overall combined feature set represented by the classification feature vector.
And, because each of the plurality of context voltage sampling window waveform feature vectors represents the image semantic feature of the context association of the output voltage local sampling window signal, the plurality of output voltage local sampling window signals are obtained by sampling the noise-reduced output voltage signal by sliding window based on sampling windows, the noise-reduced output voltage signal has a homologous data association relationship, and context association feature extraction is performed by a converter model, the plurality of context voltage sampling window waveform feature vectors have a multi-source information association relationship of a plurality of local association feature distributions corresponding to the image semantic information of the noise-reduced output voltage signal besides a neighborhood distribution relationship which is associated with each other.
Thus, in order to promote the associative distribution expression effect of the individual context voltage sampling window waveform feature vectors constituting the classification feature vector as a whole, the applicant of the present application refers to the individual context voltage sampling window waveform feature vectors, for example, denoted as V i Performing multisource information fusion pre-verification distribution evaluation optimization to obtain optimized context voltage sampling window waveform feature vector V '' i 。
Optionally, in an embodiment of the present application, the feature optimization unit 152 is configured to: carrying out multisource information fusion pre-verification distribution evaluation optimization on the context voltage sampling window waveform characteristic vectors by using the following optimization formula to obtain the optimized context voltage sampling window waveform characteristic vectors;
wherein, the optimization formula is:
wherein V is i Is the plurality of upper partsThe ith context voltage sampling window waveform characteristic vector in the context voltage sampling window waveform characteristic vectors is V j Is a j-th context voltage sampling window waveform feature vector of the plurality of context voltage sampling window waveform feature vectors,is a mean value feature vector, n is a neighborhood setting hyper-parameter, log represents a logarithmic function value based on 2,/for the neighborhood setting hyper-parameter>Representing subtraction by position, V' i Is the ith optimized context voltage sampling window waveform feature vector in the plurality of optimized context voltage sampling window waveform feature vectors.
Here, the optimization of the multisource information fusion pre-verification distribution evaluation can be used for realizing effective folding of the pre-verification information of each feature vector on the local synthesis distribution based on the quasi-maximum likelihood estimation of the feature distribution fusion robustness for the feature local collection formed by a plurality of mutually-associated neighborhood parts, and the optimization paradigm of standard expected fusion information which can be used for evaluating the internal association in the collection and the change relation between the collection is obtained through the pre-verification distribution construction under the multisource condition, so that the information expression effect of the feature vector fusion based on the multisource information association is improved. Therefore, the classification feature vector obtained by cascading the optimized context voltage sampling window waveform feature vector can have a better overall expression effect, so that the accuracy of a classification result obtained by the classifier is improved. Therefore, whether a power failure event occurs to the low-voltage side power supply can be accurately detected, normal operation of the collector is guaranteed, and safety and stability of the operation of the collector are improved.
Optionally, in another embodiment of the present application, performing waveform feature extraction and inter-waveform correlation feature extraction on the plurality of output voltage local sampling window signals to obtain a classification feature vector includes: firstly, extracting time domain features and frequency domain features of the plurality of output voltage local sampling window signals respectively to obtain a plurality of output voltage local sampling window signal feature vectors. It will be appreciated that the plurality of output voltage local sampling window signals have different characteristics in both the time and frequency domains. The time domain features mainly reflect the time sequence characteristics of the signals, including the amplitude, period, rising time, falling time, peak value and the like of the signals; the frequency domain features reflect the frequency components and frequency distribution of the signals, including power spectral density, frequency amplitude, frequency phase, frequency peak, frequency bandwidth, etc. The extraction of the time domain features and the frequency domain features is complementary when the feature extraction is performed on the plurality of output voltage local sampling window signals. Time domain features can help to understand the timing characteristics of a signal, such as rise and fall times of the signal, which are important in determining rapid changes and transient response of the signal. The frequency domain features can help to understand the frequency components and frequency distribution of the signal, such as the dominant frequency and harmonics of the signal, and are important for determining the periodicity and frequency distribution of the signal. Therefore, the time domain features and the frequency domain features of the plurality of output voltage local sampling window signals are extracted respectively, so that more comprehensive and more accurate feature vectors can be obtained and used for detecting power failure. Meanwhile, the complementarity of the time domain features and the frequency domain features can also improve the distinguishing property and the robustness of the feature vectors, so that the accuracy and the reliability of power failure detection are improved.
And then, carrying out waveform alignment on the output voltage local sampling window signals to obtain a plurality of aligned output voltage local sampling window signals. It will be appreciated that since the start times of different sampling windows may be different, it is necessary to align the sampling windows so that signals at the same location in the sampling windows may correspond. Optionally, in a sub-embodiment of the present embodiment, the plurality of output voltage local sampling window signals are waveform aligned using a cross correlation function. Here, the cross-correlation function is a function for measuring the similarity between two signals. It is calculated by comparing one signal with a lagging version of the other signal. In particular, the cross-correlation function is a dot product function between two signals, one of which is a delayed version of the other. Optionally, in another sub-embodiment of the present embodiment, the plurality of output voltage local sampling window signals are waveform aligned using a cross-correlation function.
And then calculating the waveform similarity among the plurality of aligned output voltage local sampling window signals to obtain a plurality of aligned waveform similarity. It will be appreciated that since under normal conditions the waveforms between different sampling windows should be highly similar, if anomalies are present, this will lead to increased variability in the waveforms. By calculating the waveform similarity among different sampling windows, abnormal conditions can be rapidly and accurately detected, so that measures can be timely taken to carry out fault diagnosis and repair. In addition, calculating the waveform similarity between different sampling windows can also be used for judging the stability and reliability of signals, so that the reliability and stability of the system are improved.
And then, carrying out associated feature extraction on the plurality of alignment waveform similarities. It should be appreciated that since in practical situations there may be multiple factors leading to increased waveform differences, a single waveform similarity may not fully reflect the effects of these factors. Therefore, by extracting the correlation characteristics of the plurality of alignment waveform similarities, the influence of a plurality of factors can be comprehensively considered, so that whether the output of the low-voltage side power supply of the collector is abnormal or not can be accurately judged. For example, there may be some factors that cause the variance of a certain sampling window waveform to increase, but the variance of other sampling window waveforms does not increase, in which case a single waveform similarity may not reflect an anomaly. And by extracting the correlation characteristics of the plurality of alignment waveform similarities, the waveform similarities of the plurality of sampling windows can be comprehensively considered, so that whether the output of the low-voltage side power supply of the collector is abnormal or not can be accurately judged.
Finally, a classification feature vector is generated based on the plurality of output voltage local sampling window signal feature vectors and the correlation feature. It should be understood that, since the signal feature vectors and the associated features of the local sampling windows of the multiple output voltages can only reflect part of the information, the influence of multiple factors can be comprehensively considered after the signal feature vectors and the associated features are combined, and therefore whether the output of the low-voltage side power supply of the collector is abnormal or not can be accurately judged. And combining the obtained correlation characteristic and the signal characteristic vectors of the plurality of output voltage local sampling windows to generate a final classification characteristic vector, so that the accuracy and the reliability of fault diagnosis can be improved. Meanwhile, as the feature vector has higher dimension, the relation among different features can be reflected better, so that the efficiency and the precision of fault diagnosis are improved.
The power-down detection module 160 is configured to pass the classification feature vector through a classifier to obtain a classification result, where the classification result is used to indicate whether a power-down event occurs. It should be understood that in the technical solution of the present application, the labels of the classifier include that a power-down event (first label) occurs, and that no power-down event (second label) occurs, where the classifier determines to which classification label the classification feature vector belongs through a soft maximum function. It should be noted that the first tag p1 and the second tag p2 do not include a manually set concept, and in fact, during the training process, the computer model does not have a concept of "whether a power failure event occurs", which is only two kinds of classification tags, and the probability that the output feature is under the two classification tags, that is, the sum of p1 and p2 is one. Therefore, the classification result of whether the power-down event occurs is actually converted into a classified probability distribution conforming to the natural rule through classifying the tag, and the physical meaning of the natural probability distribution of the tag is essentially used instead of the language text meaning of whether the power-down event occurs. It should be understood that, in the technical scheme of the present application, the classification label of the classifier is a detection evaluation label for detecting whether a power failure event occurs, so after the classification result is obtained, whether the power failure event occurs to the low-voltage side power supply can be detected based on the classification result, thereby ensuring the normal operation of the collector.
Optionally, in an embodiment of the present application, the power-down detection module 160 is configured to: processing the classification feature vector by using the classifier according to the following classification formula to obtain the classification result used for indicating whether a power-down event occurs;
wherein, the classification formula is: o=softmax { (W) c ,B c ) V, where V represents the classification of the classificationFeature vector, W c Is a weight matrix, B c Representing a bias vector, softmax representing a normalized exponential function, and O representing the classification result for indicating whether a power down event has occurred.
In summary, the power failure detection system of the collector provided by the application can accurately detect whether a power failure event occurs to a low-voltage side power supply, thereby ensuring the normal operation of the collector and improving the safety and stability of the operation of the collector.
Fig. 4 is a schematic flow chart of a collector power failure detection method according to an embodiment of the present application. Fig. 5 is a schematic diagram of a model architecture of a collector power failure detection method according to an embodiment of the present application. As shown in fig. 4 and 5, the method includes: s110, obtaining an output voltage signal output by a low-voltage side power supply; s120, carrying out noise reduction processing on the output voltage signal to obtain a noise-reduced output voltage signal; s130, sliding window type sampling based on a sampling window is carried out on the noise-reduced output voltage signal so as to obtain a plurality of output voltage local sampling window signals; s140, the output voltage local sampling window signals are respectively passed through a voltage waveform characteristic extractor based on a convolutional neural network model to obtain a plurality of voltage sampling window waveform characteristic vectors; s150, passing the waveform feature vectors of the voltage sampling windows through a waveform inter-correlation feature extractor based on a converter model to obtain classification feature vectors; and S150, passing the classification feature vector through a classifier to obtain a classification result, wherein the classification result is used for indicating whether a power-down event occurs.
Here, it will be understood by those skilled in the art that the specific operations of the respective steps in the above-described collector power failure detection method have been described in detail in the above description of the collector power failure detection system with reference to fig. 1 to 3, and thus, repetitive descriptions thereof will be omitted.
The embodiment of the application also provides a chip system, which comprises at least one processor, and when the program instructions are executed in the at least one processor, the method provided by the embodiment of the application is realized.
The embodiment of the application also provides a computer storage medium, on which a computer program is stored, which when executed by a computer causes the computer to perform the method of the above-described method embodiment.
The present application also provides a computer program product comprising instructions which, when executed by a computer, cause the computer to perform the method of the method embodiment described above.
In the above embodiments, it may be implemented in whole or in part by software, hardware, firmware, or any other combination. When implemented in software, may be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When loaded and executed on a computer, produces a flow or function in accordance with embodiments of the present application, in whole or in part. The computer may be a general purpose computer, a special purpose computer, a computer network, or other programmable apparatus. The computer instructions may be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium, for example, the computer instructions may be transmitted from one website, computer, server, or data center to another website, computer, server, or data center by a wired (e.g., coaxial cable, fiber optic, digital subscriber line (digital subscriber line, DSL)) or wireless (e.g., infrared, wireless, microwave, etc.). The computer readable storage medium may be any available medium that can be accessed by a computer or a data storage device such as a server, data center, etc. that contains an integration of one or more available media. The usable medium may be a magnetic medium (e.g., a floppy disk, a hard disk, a magnetic tape), an optical medium (e.g., a digital video disc (digital video disc, DVD)), or a semiconductor medium (e.g., a Solid State Disk (SSD)), or the like.
Those of ordinary skill in the art will appreciate that the various illustrative elements and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware, or combinations of computer software and electronic hardware. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the solution. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present invention.
In the several embodiments provided by the present invention, it should be understood that the disclosed systems, devices, and methods may be implemented in other manners. For example, the apparatus embodiments described above are merely illustrative, e.g., the division of the units is merely a logical function division, and there may be additional divisions when actually implemented, e.g., multiple units or components may be combined or integrated into another system, or some features may be omitted or not performed. Alternatively, the coupling or direct coupling or communication connection shown or discussed with each other may be an indirect coupling or communication connection via some interfaces, devices or units, which may be in electrical, mechanical or other form.
The units described as separate units may or may not be physically separate, and units shown as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
In addition, each functional unit in the embodiments of the present invention may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit.
Claims (6)
1. A collector power down detection system, comprising:
the signal acquisition module is used for acquiring an output voltage signal output by the low-voltage side power supply;
the signal noise reduction module is used for carrying out noise reduction processing on the output voltage signal to obtain a noise-reduced output voltage signal;
the sampling module is used for carrying out sliding window type sampling based on the sampling window on the noise-reduced output voltage signal so as to obtain a plurality of output voltage local sampling window signals;
the voltage waveform characteristic extraction module is used for respectively passing the plurality of output voltage local sampling window signals through a voltage waveform characteristic extractor based on a convolutional neural network model to obtain a plurality of voltage sampling window waveform characteristic vectors;
The waveform characteristic global correlation module is used for enabling the waveform characteristic vectors of the voltage sampling windows to pass through a waveform inter-correlation characteristic extractor based on a converter model to obtain classification characteristic vectors; and
the power-down detection module is used for enabling the classification feature vector to pass through a classifier to obtain a classification result, wherein the classification result is used for indicating whether a power-down event occurs or not;
the waveform characteristic global association module comprises:
a waveform correlation feature extraction unit, configured to pass the plurality of voltage sampling window waveform feature vectors through the inter-waveform correlation feature extractor based on the converter model to obtain a plurality of context voltage sampling window waveform feature vectors;
the characteristic optimization unit is used for carrying out multisource information fusion pre-verification distribution evaluation optimization on the waveform characteristic vectors of the context voltage sampling windows so as to obtain the waveform characteristic vectors of the context voltage sampling windows after the optimization;
the cascade fusion unit is used for cascading the optimized context voltage sampling window waveform characteristic vectors to obtain the classification characteristic vectors;
wherein the waveform-associated feature extraction unit includes:
the characteristic vector arrangement subunit is used for one-dimensionally arranging the waveform characteristic vectors of the voltage sampling windows to obtain a waveform global characteristic vector;
A vector multiplication subunit, configured to calculate a product between the global waveform feature vector and a transpose vector of each of the plurality of voltage sampling window waveform feature vectors to obtain a plurality of self-attention correlation matrices;
the incidence matrix processing subunit is used for respectively carrying out standardization processing on each self-attention incidence matrix in the plurality of self-attention incidence matrixes so as to obtain a plurality of standardized self-attention incidence matrixes;
the probability unit is used for obtaining a plurality of probability values through classification functions by each normalized self-attention correlation matrix in the normalized self-attention correlation matrices; and
and the probability applying unit is used for weighting each voltage sampling window waveform characteristic vector in the plurality of voltage sampling window waveform characteristic vectors by taking each probability value in the plurality of probability values as a weight so as to obtain the plurality of context voltage sampling window waveform characteristic vectors.
2. The collector power down detection system of claim 1, wherein the voltage waveform feature extraction module is configured to: each layer of the voltage waveform characteristic extractor based on the convolutional neural network model is used for respectively carrying out input data in forward transfer of the layer:
Performing convolution processing on the input data based on convolution check to generate a convolution feature map;
performing global average pooling processing based on a feature matrix on the convolution feature map to generate a pooled feature map; and
non-linear activation is carried out on the characteristic values of all the positions in the pooled characteristic map so as to generate an activated characteristic map;
the output of the last layer of the voltage waveform characteristic extractor based on the convolutional neural network model is the waveform characteristic vector of each voltage sampling window in the waveform characteristic vectors of the voltage sampling windows, the input of the second layer to the last layer of the voltage waveform characteristic extractor based on the convolutional neural network model is the output of the last layer, and the input of the voltage waveform characteristic extractor based on the convolutional neural network model is the output of each output voltage local sampling window in the partial sampling window signals of the output voltage.
3. The collector power down detection system of claim 2, wherein the feature optimization unit is configured to: carrying out multisource information fusion pre-verification distribution evaluation optimization on the context voltage sampling window waveform characteristic vectors by using the following optimization formula to obtain the optimized context voltage sampling window waveform characteristic vectors;
Wherein, the optimization formula is:
，
wherein,is the +.f. of the waveform eigenvectors of the multiple context voltage sampling windows>Waveform characteristic vector of each context voltage sampling window, < > and the like>Is the +.f. of the waveform eigenvectors of the multiple context voltage sampling windows>Waveform characteristic vector of each context voltage sampling window, < > and the like>Is the mean feature vector, ++>Setting up hyper-parameters for the neighborhood, log representing the logarithmic function value based on 2, ⊖ representing the per-position subtraction,/->Is the +.f. of the waveform eigenvectors of the multiple optimized context voltage sampling windows>And the waveform characteristic vector of the optimized context voltage sampling window.
4. The collector power down detection system of claim 3 wherein the power down detection module is configured to: processing the classification feature vector by using the classifier according to the following classification formula to obtain the classification result used for indicating whether a power-down event occurs;
wherein, the classification formula is:wherein->Representing the classification feature vector, +.>Is a weight matrix>Representing the bias vector +_>Representing normalized exponential function, ++>And representing the classification result for representing whether the power-down event occurs.
5. The collector power failure detection method is characterized by comprising the following steps of:
Obtaining an output voltage signal output by a low-voltage side power supply;
carrying out noise reduction treatment on the output voltage signal to obtain a noise-reduced output voltage signal;
sliding window type sampling based on sampling windows is carried out on the noise-reduced output voltage signals so as to obtain a plurality of output voltage local sampling window signals;
the output voltage local sampling window signals are respectively passed through a voltage waveform characteristic extractor based on a convolutional neural network model to obtain a plurality of voltage sampling window waveform characteristic vectors;
passing the plurality of voltage sampling window waveform feature vectors through a converter model-based inter-waveform correlation feature extractor to obtain classification feature vectors; and
the classification feature vector passes through a classifier to obtain a classification result, wherein the classification result is used for indicating whether a power failure event occurs or not;
wherein passing the plurality of voltage sampling window waveform feature vectors through a converter model-based inter-waveform correlation feature extractor to obtain a classification feature vector comprises:
passing the plurality of voltage sampling window waveform feature vectors through the inter-waveform correlation feature extractor based on the converter model to obtain a plurality of context voltage sampling window waveform feature vectors;
Carrying out multisource information fusion pre-verification distribution evaluation optimization on the context voltage sampling window waveform feature vectors to obtain a plurality of optimized context voltage sampling window waveform feature vectors;
cascading the optimized context voltage sampling window waveform feature vectors to obtain the classification feature vectors;
wherein passing the plurality of voltage sampling window waveform feature vectors through the inter-waveform correlation feature extractor based on the converter model to obtain a plurality of context voltage sampling window waveform feature vectors comprises:
one-dimensional arrangement is carried out on the waveform characteristic vectors of the voltage sampling windows so as to obtain a waveform global characteristic vector;
calculating the product between the waveform global feature vector and the transpose vector of each of the plurality of voltage sampling window waveform feature vectors to obtain a plurality of self-attention correlation matrices;
respectively carrying out standardization processing on each self-attention correlation matrix in the plurality of self-attention correlation matrices to obtain a plurality of standardized self-attention correlation matrices;
each normalized self-attention correlation matrix in the normalized self-attention correlation matrices is subjected to a classification function to obtain a plurality of probability values; and
And weighting each voltage sampling window waveform characteristic vector in the plurality of voltage sampling window waveform characteristic vectors by taking each probability value in the plurality of probability values as a weight so as to obtain a plurality of context voltage sampling window waveform characteristic vectors.
6. The collector power down detection method of claim 5, wherein performing multisource information fusion pre-test distribution evaluation optimization on the plurality of context voltage sampling window waveform feature vectors to obtain a plurality of optimized context voltage sampling window waveform feature vectors comprises: carrying out multisource information fusion pre-verification distribution evaluation optimization on the context voltage sampling window waveform characteristic vectors by using the following optimization formula to obtain the optimized context voltage sampling window waveform characteristic vectors;
wherein, the optimization formula is: ，
wherein,is the +.f. of the waveform eigenvectors of the multiple context voltage sampling windows>Waveform characteristic vector of each context voltage sampling window, < > and the like>Is the +.f. of the waveform eigenvectors of the multiple context voltage sampling windows>Waveform characteristic vector of each context voltage sampling window, < > and the like>Is the mean feature vector, ++>Setting up hyper-parameters for the neighborhood, log representing the logarithmic function value based on 2, ⊖ representing the per-position subtraction,/- >Is the +.f. of the waveform eigenvectors of the multiple optimized context voltage sampling windows>And the waveform characteristic vector of the optimized context voltage sampling window.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310710148.3A CN116773894B (en) | 2023-06-15 | 2023-06-15 | Collector power failure detection system and method thereof |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310710148.3A CN116773894B (en) | 2023-06-15 | 2023-06-15 | Collector power failure detection system and method thereof |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116773894A CN116773894A (en) | 2023-09-19 |
| CN116773894B true CN116773894B (en) | 2023-12-15 |
Family
ID=87995745
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310710148.3A Active CN116773894B (en) | 2023-06-15 | 2023-06-15 | Collector power failure detection system and method thereof |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116773894B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117375237A (en) * | 2023-10-20 | 2024-01-09 | 浙江日新电气有限公司 | Substation operation and maintenance method and system based on digital twin technology |
| CN117614784A (en) * | 2023-11-15 | 2024-02-27 | 浙江恒业电子股份有限公司 | Wireless communication module based on carrier wave |
Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108151197A (en) * | 2017-12-29 | 2018-06-12 | 浙江中来技术有限公司 | A kind of method for managing power supply of air cleaning system and atomizer |
| CN108649939A (en) * | 2018-04-16 | 2018-10-12 | 芯原微电子(上海)有限公司 | Power sense circuit and method |
| CN109034277A (en) * | 2018-09-17 | 2018-12-18 | 吉林大学 | Power Quality Disturbance Classification Method based on multi-feature fusion and system |
| CN109444643A (en) * | 2018-12-20 | 2019-03-08 | 武汉海王科技有限公司 | A kind of single-phase sinusoidal signal power down rapid detection method and system |
| CN110907688A (en) * | 2018-09-17 | 2020-03-24 | 宁德时代新能源科技股份有限公司 | Power failure detection circuit and control method thereof |
| US11038520B1 (en) * | 2020-04-15 | 2021-06-15 | International Business Machines Corporation | Analog-to-digital conversion with reconfigurable function mapping for neural networks activation function acceleration |
| CN113203914A (en) * | 2021-04-08 | 2021-08-03 | 华南理工大学 | Underground cable early fault detection and identification method based on DAE-CNN |
| CN114563753A (en) * | 2021-04-12 | 2022-05-31 | 正泰集团研发中心(上海)有限公司 | Power failure detection method, device and equipment for electric energy meter and computer readable storage medium |
| CN115099285A (en) * | 2022-07-12 | 2022-09-23 | 绍兴九樱纺织品有限公司 | Intelligent detection method and system based on neural network model |
| WO2023019415A1 (en) * | 2021-08-17 | 2023-02-23 | 华为技术有限公司 | Power-off detection method and related device |
| CN115857651A (en) * | 2021-09-24 | 2023-03-28 | 英特尔公司 | Method and apparatus for external display power down detection |
| CN116125338A (en) * | 2022-11-28 | 2023-05-16 | 中国船舶集团有限公司第七〇九研究所 | Power grid voltage power-down detection circuit and power-down detection method |
-
2023
- 2023-06-15 CN CN202310710148.3A patent/CN116773894B/en active Active
Patent Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108151197A (en) * | 2017-12-29 | 2018-06-12 | 浙江中来技术有限公司 | A kind of method for managing power supply of air cleaning system and atomizer |
| CN108649939A (en) * | 2018-04-16 | 2018-10-12 | 芯原微电子(上海)有限公司 | Power sense circuit and method |
| CN109034277A (en) * | 2018-09-17 | 2018-12-18 | 吉林大学 | Power Quality Disturbance Classification Method based on multi-feature fusion and system |
| CN110907688A (en) * | 2018-09-17 | 2020-03-24 | 宁德时代新能源科技股份有限公司 | Power failure detection circuit and control method thereof |
| CN109444643A (en) * | 2018-12-20 | 2019-03-08 | 武汉海王科技有限公司 | A kind of single-phase sinusoidal signal power down rapid detection method and system |
| US11038520B1 (en) * | 2020-04-15 | 2021-06-15 | International Business Machines Corporation | Analog-to-digital conversion with reconfigurable function mapping for neural networks activation function acceleration |
| CN113203914A (en) * | 2021-04-08 | 2021-08-03 | 华南理工大学 | Underground cable early fault detection and identification method based on DAE-CNN |
| CN114563753A (en) * | 2021-04-12 | 2022-05-31 | 正泰集团研发中心(上海)有限公司 | Power failure detection method, device and equipment for electric energy meter and computer readable storage medium |
| WO2023019415A1 (en) * | 2021-08-17 | 2023-02-23 | 华为技术有限公司 | Power-off detection method and related device |
| CN115857651A (en) * | 2021-09-24 | 2023-03-28 | 英特尔公司 | Method and apparatus for external display power down detection |
| CN115099285A (en) * | 2022-07-12 | 2022-09-23 | 绍兴九樱纺织品有限公司 | Intelligent detection method and system based on neural network model |
| CN116125338A (en) * | 2022-11-28 | 2023-05-16 | 中国船舶集团有限公司第七〇九研究所 | Power grid voltage power-down detection circuit and power-down detection method |
Non-Patent Citations (1)
| Title |
|---|
| 一种智能电能表掉电检测模块的电路设计;高希栋;《机电信息》;第2018卷(第15期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116773894A (en) | 2023-09-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN116773894B (en) | Collector power failure detection system and method thereof | |
| CN111238814B (en) | Rolling bearing fault diagnosis method based on short-time Hilbert transform | |
| US20030187621A1 (en) | Method, computer program, and system for automated real-time signal analysis for detection, quantification, and prediction of signal changes | |
| CN110579690B (en) | Positioning method for small current ground fault section of power distribution network | |
| CN109085614B (en) | GNSS interference feature identification method and system based on time-frequency spectrum value smoothing and segmentation | |
| CN104569694A (en) | Electric signal feature extraction and recognition system oriented to aircraft flying process | |
| CN115376526A (en) | Power equipment fault detection method and system based on voiceprint recognition | |
| CN113257249A (en) | Power equipment fault diagnosis method, device and equipment based on voiceprint recognition | |
| CN115081331A (en) | Wind turbine generator running state abnormity detection method based on state parameter reconstruction error | |
| Leung et al. | Reconstruction of FRI signals using deep neural network approaches | |
| Upendar et al. | Discrete wavelet transform and probabilistic neural network based algorithm for classification of fault on transmission systems | |
| Hoffmann et al. | Techniques for the diagnosis of oscillatory transients resulting from capacitor bank switching in medium voltage distribution systems | |
| CN112182490B (en) | Reactor state diagnosis method and system | |
| Conru et al. | Time-frequency detection using Gabor filter bank and Viterbi based grouping algorithm | |
| CN107578016B (en) | Sparse representation-based automatic residual current waveform identification method | |
| CN112151067B (en) | Digital audio tampering passive detection method based on convolutional neural network | |
| CN113791318A (en) | Partial discharge fault identification method and system based on multispectral detection array | |
| CN115902528B (en) | Method for identifying oscillation and short-circuit faults of direct-current traction network | |
| CN116189711B (en) | Transformer fault identification method and device based on acoustic wave signal monitoring | |
| CN112033656A (en) | Mechanical system fault detection method based on broadband spectrum processing | |
| CN112697270A (en) | Fault detection method and device, unmanned equipment and storage medium | |
| CN116230013A (en) | Transformer fault voiceprint detection method based on x-vector | |
| CN106096494B (en) | A kind of traveling wave fault based on information fusion framework and interference recorder data discrimination method | |
| CN113505654A (en) | Bearing health state identification method based on multi-view attention network | |
| CN117290670B (en) | Transformer bushing insulation state estimation method based on enhancement filter algorithm |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |